Markerless augmented reality remote laboratory for computer networks education in hybrid learning: a quasi-experimental study

Abstract

This study evaluates a markerless augmented reality (AR)-based remote laboratory for supporting hybrid learning in a computer networks course. Conventional remote laboratory approaches based on two-dimensional network simulations may provide flexible access but offer limited spatial interaction with networking components. To address this limitation, a markerless AR remote laboratory was developed using the Niantic Lightship Geospatial Browser, enabling students to place, connect, and configure virtual routers, switches, and end devices within their physical surroundings. A quasi-experimental design was conducted over one academic semester (14 weeks) with 120 undergraduate students, comprising an experimental group using the AR-based remote laboratory (n=60) and a control group using a conventional two-dimensional network simulation environment (n=60). Learning outcomes, usability, engagement, interaction logs, and qualitative feedback were collected and analyzed. The experimental group achieved higher post-test scores than the control group (82.45±6.31 vs. 74.18±7.02, p<0.001), with a large between-group effect (Cohen’s d=1.25). The AR group also reported higher engagement and learning confidence, while interaction logs indicated shorter task completion times and fewer configuration attempts. These findings suggest that markerless AR can provide a promising approach for supporting spatial interaction and experiential quality in remote computer networks laboratory activities within hybrid learning environments.